Research Area:  Machine Learning
To proactively fulfill multiple stakeholders needs in the engineering solution design process, the knowledge recommendation approach is adopted as a key element in the knowledge management system. Nevertheless, most existing knowledge recommendation approaches cannot simultaneously meet the higher standard of in-context accuracy and diversity. To address the issue, this paper proposes a context-aware diversity-oriented knowledge recommendation approach, thereby assisting stakeholders to accomplish engineering solution design in a smarter manner. Three diversity concerns, namely item-diversity, context-diversity, and user-diversity are addressed by semantic-based content analysis, context definition and awareness, and user profile modeling, respectively. Hence, the proposed approach not only maximizes the diversity of the recommended knowledge but also guarantees its accuracy under multiple problem-solving contexts. Moreover, a practical engineering solution design case on a Smart 3D printer platform is conducted, to validate the efficacy of the proposed approach in providing usable and diverse knowledge items. It is anticipated this work can provide useful insights to practitioners in their knowledge-based engineering solution design process.
Author(s) Name:  Xinyu Li,Chun-Hsien Chen,Pai Zheng,Zuhua Jiang,Linke Wang
Journal name:  Knowledge-Based Systems
Publisher name:  Elsevier
Volume Information:  Volume 215, 5 March 2021, 106739 Knowledge-Based Systems
Paper Link:   https://www.sciencedirect.com/science/article/abs/pii/S0950705121000022